SKILLEMALL.ai

FC download-video

Download videos from 1000+ websites (YouTube, Bilibili, Twitter/X, TikTok, Vimeo, Instagram, Twitch, etc.) using yt-dlp. Use this skill whenever a user shares a video URL, asks to save or download a video, wants to extract audio from an online video, needs a specific quality like 1080p or 4K, or mentions downloading a playlist. Also trigger on "下载视频", "保存视频", "提取音频", or any URL from a supported video platform.

Not recommendedcritical or high security findings · low grade F
ClawHub Agent Skills author: Pengfei Ni v1.0.1 MIT-0 4 files body ≈ 502 tokens Open the sourceclawhub.ai analyzed 3 d ago

As a process C 57/100 · Has gaps — weak spots: steps, result and completion, consistency

ReferenceYouTubeMedia and videotype and topics are labelled automatically from the skill text
JSON
Technical rating
F
36/100
safety, quality, tests
Safety 60%
1
Quality 40%
88
Run on models
none yet
Process rating
C
57/100
Has gaps
Steps w 15
0
Result and completion w 14
0
Progress reporting w 2
0
the three weakest of ten parameters · all ten

What is at stake

The skill contains fragments that, in the wrong hands, cost money or data. Below: what the installer risks and what the author should do.

Exfiltration
If you install

The instructions or scripts send data out: environment variables, keys, file contents, chat history. You may never notice, because the agent performs the upload as "part of the task".

For the author

If the upload is not needed for the task, remove it: catalogs flag such skills and delist them. If it is needed, name the destination explicitly, say what leaves the machine, and give the user a switch.

How to improve

  1. Address the high-severity findings: each costs 18 safety points. If one is a false positive, add the rule id to guard.allow in spec.yaml.
For the model run — optional
  • Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
  • A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.

Guard findings · 12

  • high Exfiltration intent-browser-credential-store references/platform-tips.md:12
    Accesses a browser credential / cookie store
    yt-dlp --cookies-from-browser chrome "VIDEO_URL"
  • high Exfiltration intent-browser-credential-store references/platform-tips.md:34
    Accesses a browser credential / cookie store
    yt-dlp --cookies-from-browser chrome "VIDEO_URL"
  • high Exfiltration intent-browser-credential-store SKILL.md:45
    Accesses a browser credential / cookie store
    python3 scripts/download.py "URL" --cookies chrome    # Use browser cookies
Medium and low: 9
  • medium Exfiltration intent-browser-credential-store references/platform-tips.md:43
    Accesses a browser credential / cookie store (the skill's own vendor host)
    yt-dlp --cookies-from-browser chrome "https://www.bilibili.com/video/BV..."
    vendor-host
  • medium Exfiltration intent-browser-credential-store references/platform-tips.md:67
    Accesses a browser credential / cookie store (the skill's own vendor host)
    yt-dlp --cookies-from-browser chrome "https://twitter.com/user/status/123456789"
    vendor-host
  • medium Exfiltration intent-browser-credential-store references/platform-tips.md:85
    Accesses a browser credential / cookie store (the skill's own vendor host)
    yt-dlp --cookies-from-browser chrome "https://www.instagram.com/p/ABC123/"
    vendor-host
  • medium Exfiltration intent-browser-credential-store scripts/download.py:63
    Accesses a browser credential / cookie store (quoted — discussed, not commanded)
    use_cookies: Browser to extract cookies from (chrome, firefox, etc.)
    quoted
  • medium Exfiltration intent-browser-credential-store scripts/download.py:119
    Accesses a browser credential / cookie store (quoted — discussed, not commanded)
    print("  2. Try with cookies: python3 download.py URL --cookies chrome")
    quoted
  • medium Exfiltration intent-browser-credential-store scripts/download.py:132
    Accesses a browser credential / cookie store (quoted — discussed, not commanded)
    %(prog)s "https://www.bilibili.com/video/BV..." --cookies chrome
    quoted
  • medium Exfiltration intent-browser-credential-store scripts/download.py:168
    Accesses a browser credential / cookie store (quoted — discussed, not commanded)
    help="Browser to extract cookies from (chrome, firefox, safari, edge)",
    quoted
  • medium Exfiltration intent-browser-credential-store SKILL.md:69
    Accesses a browser credential / cookie store (documentation table row)
    | "Sign in required" or age-restricted | Add `--cookies chrome` to use browser session |
    table
  • medium Exfiltration intent-browser-credential-store SKILL.md:70
    Accesses a browser credential / cookie store (documentation table row)
    | Only low quality available | Update yt-dlp (`brew upgrade yt-dlp`), then try with `--cookies chrome` |
    table

Files scanned: 4. Evidence is masked. Grey chips explain why severity was lowered.

Against the Agent Skills spec

✓ No remarks against the Agent Skills spec

Process rating: all ten parameters 57/100

  • 0Steps. Prose only: no discrete steps
  • 0Result and completion. Does not say what the result is
  • 0Progress reporting. Says nothing while it works
  • 40Consistency. Frontmatter name (download-video) differs from the folder (feiskyer-download-video)
  • 70When it triggers. States when to use, but not when not to
  • 70Inputs and preconditions. Inputs and preconditions are listed
  • 100Tools and files. No external tools needed
  • 100Failures and branches. 1 branches, has a failure section
  • 100Execution cost. Instruction body is 502 tokens
  • 100Running it twice. No mutating operations

Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.

Quality signals

  • +5Description has no quoted example phrases that should trigger the skill
  • +4Description does not say when NOT to use the skill (false activations)
  • +3No numbered steps or checklist
  • +3Output format is not stated: the model decides each time
  • +1No license
  • +2Single-language instructions
  • +3Description length 413: enough signal without eating the budget
  • +4Structure: 6 headings
  • +4Has examples (5 code blocks)
  • +4Reference files are cited in the instructions (1 of 1)
  • +3All 1 scripts are documented

Quality base 70; lint remarks subtract, signals add up to 100. Result: 88.

External checks

ClawHub: suspicious
This video-downloader skill appears purpose-related, but it can use browser login cookies for authenticated downloads without sufficiently clear scoping or consent.
LLM: suspicious (medium) · VirusTotal: · 29 May 2026